DESCRIPTION: MCF-7 breast cancer cells are estrogen-responsive in vitro, estrogen-dependent in vivo, and are sensitive to treatment with antiestrogens. They are also poorly metastatic in athymic nude mice. Overexpression of either of two angiogenic growth factors, FGF-1 or FGF-4 has strong effects on the tumorigenic and metastatic phenotypes. FGF-1 or FGF-4 transfected cells acquire the ability to form progressive tumors in ovariectomized nude mice without estrogen supplementation or in mice that receive the antiestrogen tamoxifen. Both types of transfectants also show a dramatically increased ability to form micrometastases. FGF signal transduction operating inhuman breast tumors is likely since a majority of tumors express mRNA for FGF receptors and FGF-1 and FGF-2. These results point out the possible importance of the paracrine and/or autocrine effects elicited by FGF and FGFR expression in human breast cancer. The proposed experiments will use transfection techniques to study the mechanisms responsible for the phenotypic changes caused by FGF-1 and FGF-4 overexpression. They are designed to clarify the relative contributions of the possible autocrine, paracrine, or intracrine effects of FGF and other growth factors with angiogenic potential in the progression of human breast cancer. Aim 1 will determine if cells at an earlier stage in the breast cancer progression pathway can acquire tumorigenic and metastatic phenotypes as a result of either FGF or FGFR overexpression. Immortalized MCF-10A breast epithelial cells will be transfected with FGF-1, FGF-4, and FGFR expression vectors. Aim 2 will determine if the paracrine effects of FGF expression alone are sufficient to result in increased tumorigenicity, antiestrogen insensitivity, or increased metastatic potential. MCF-7 cells overexpressing FGF-1 or FGF-4 will be retransfected with an expression vector for a truncated FGF receptor capable of acting in a dominant- negative fashion. This vector will be used to turn off FGF signal transduction within the human breast tumor cells without interfering with their ability to secrete these factors. This will allow a continuation of any paracrine effects of FGF production in the nude mouse and indicate whether intracrine or autocrine FGF loops are required for the phenotypes observed in FGF overexpressing cells. Aim 3 will explore the contribution of neoangiogenesis to the tumorigenic, metastatic and antiestrogen- resistant phenotypes. MCF-7 cells will be transfected with a vector directing the overexpression of vascular endothelial cell growth factor (VEGF). This factor does not have a direct mitogenic effect on breast cancer cells and is a specific mitogen for endothelial cells. This transfection will test the hypothesis that the development of a supporting vascular network resulting from the paracrine effects of VEGF expression will confer VEGF- overexpressing cells with the same in vivo phenotypes observed with cells overexpressing FGFs. MCF-7 cells overexpressing the c-erb-2 transmembrane tyrosine kinase as a result of transfection are capable of developing tumors in untreated- and tamoxifen-treated mice but these tumors fail to grow progressively and do not form metastases. In Aim 4 MCF-7 cells will be transfected with VEGF and c-erbB-2 expression vectors. This double transfection will determine if acquisition of increased tumorigenic and metastatic phenotypes requires the combination of a direct mitogenic effect resulting from activation of growth factor signal transduction pathways and increased vascularization resulting from expression of an angiogenic growth factor. The results of these studies may suggest that approaches targeted at interfering with neoangiogenesis or FGF signal transduction may have therapeutic value in the treatment of human breast cancer.